3103 Network Inductors DCR

[80n]

Hi, I'm trying to build a 3103 network from scratch.
Schematic here:

http://www.jblproservice.com/pdf/Net...%20Network.pdf

I really need the DCR value of both inductances 57530 (0.4mH) & 57531 (1.2mH) which are found in this network.

Anybody can help ?

[Baron030]

Hi 80n

I don’t know the DCR values on those coils. But, since the schematic shows a symbol for iron core coils then the DCR values should be fairly low. And maybe someone could chime with a wire gauges used on those coils, which could give you a clue as to the exact DCR values.

When I was rebuilding some N2400 networks, I found that the iron core coil had 20 gauge leads. So, unless the 3103 network is build for some really high power applications then I would seriously doubt that the wire used on the 1.2mH coil would be greater then 18 gauge. And I would think that the wire used in the 0.4mH coil is only 20 gauge.

So, if I were building a pair 3103 networks for my own use then I would go with 1.2 mH – 14 or 12 gauge air core coils and a 0.4mH - 18 gauge air core coils. The advantage of using air cores coils is that they do not saturate like iron core coils.

Baron030

[4313B]

I really need the DCR value of both inductances 57530 (0.4mH) & 57531 (1.2mH) which are found in this network.

In this particular instance you really don't need to know. Anything under 0.5 to 0.6 ohms will be fine.

The advantage of using air cores coils is that they do not saturate like iron core coils.

Yep.

[80n]

Actually, I already "hastily" bought some Jantzen 20 AWG air core inductors :

- 1.2 mH / 0.8 ohm DCR
- 0.4 mH / 0.43 ohm DCR

Do you confirm they are not appropriate ?

A bit of context : the goal here is to upgrade my Decade L16 crossovers while not spending big money

thanks !

[Baron030]

Actually, I already "hastily" bought some Jantzen 20 AWG air core inductors:
- 1.2 mH / 0.8 ohm DCR
- 0.4 mH / 0.43 ohm DCR
Do you confirm they are not appropriate ?
A bit of context: the goal here is to upgrade my Decade L16 crossovers while not spending big money
thanks!

Well, the Jantzen 0.4mH – 20 awg coils will work just file for the HF drivers. But, the 1.2mH - 20 AWG coils for the woofer is pretty thin. I don’t think it will have much effect on the crossover point. But, the high DCR will impact the damping on woofer, 0.8 ohms is a lot of resistance.

Try contacting your supplier and see if you can exchange the 1.2mh - 20 AWG coils for 1.2mH – 15 AWG Jantzen air core or their p-core coils.

Baron030

[80n]

Thanks a lot, Baron030. That was the most useful post. I will follow your first idea of going with those (and try to sell my 20 gauge on ebay):

• 1.2 mH – 14 or 12 gauge air core coils
• 0.4mH - 18 gauge air core coils (those are cheap ones, anyway)

I have one last concern : My source is a Marantz 1070 amplifier. It is rated 35 W RMS per Channel. Here is my 3103 Network ""redesign"" :



Won't some parts of the new crossover get hotter than the L16 original ? I'm thinking of putting it outside the box and attach it to the backpanel available for everyone to see :

But what are the currents and voltages (and also component temperatures) involved? I want to be sure it is not dangerous (and also I am curious of those values), even if the goal is ultimately to cover this up with transparent plexiglass.

[Baron030]

Won't some parts of the new crossover get hotter than the L16 original? I'm thinking of putting it outside the box and attach it to the back panel available for everyone to see:
But what are the currents and voltages (and also component temperatures) involved? I want to be sure it is not dangerous (and also I am curious of those values), even if the goal is ultimately to cover this up with transparent Plexiglas.

What? You are worried about something getting hot?
Nope, you have nothing to worry about.

The 5 ohms 10 watt resistor is part of a noble network and it is not going to get hot.
And the 8 ohms L-pad has a 15 watt power rating (JBL part # 10285) and it’s not going to get hot either.

Let’s put it this way, if you were to pump enough power into your system to get any of these network components hot then your L16 drivers would be toast. We are talking smoking voice coils here. So, short of plugging your speakers into a wall outlet, I don’t see how a 3103 network is going to get hot enough to melt plexiglas.
Baron030

[80n]

Oh, nice.
So, what about the voltages and currents involved, are they safe too? If so, why use capacitors which are rated for 250V ? (or maybe I'm mixing everything)

Thanks to educate me

[80n]

Also, I need to find 5ohm, 10W resistors (part 13203). Should I take classic wirewound (which was probably the original), or use Metal Oxide Film resistors as I read everywhere :

"Metal Oxide Film resistors are preferred by many audio circuit designers due to their low inductance. Metal oxide films do not burn, thus withstanding heat better. They are mechanically and electrically very stable and they have a fast impulse response, making them suitable for high end audio equipment."

[Baron030]

Hi 80n
The 5 ohm 10 watt resistor is part of a zobel network which compensates for the woofer’s voice coil inductance. So, you can use a standard 10 Watt wire wound resistor in this application.
The amount of inductance in a standard 10 Watt wire wound resistor is so small that it will have no impact of how the zobel network performs. Besides, it’s down stream from the 1.2mH coil. So there are no high frequency signals present for resistor inductance to be an issue.
Baron030

[80n]

Wow. One more time, that was very instructive. I guess I have now all the info to dive into this project. I'll keep this thread updated once I gather all the necessary parts and equipment .. Which might take time.

Thank you very much, Baron030


Also, if someone need a pair of Jantzen 20 AWG air core inductors , I have those 1.2 mH / 0.8 ohm DCR and 0.4 mH / 0.43 ohm DCR that I am willing to give out for free (except postage). Just PM me if you are interested (and I leave in Belgium)

[80n]

And here is the beast : completed custom made 3103 Network.



Took me some time, but here it is, and it sounds great so far.
The white blobs are silicon. Components are directly connected. Inductors are secured with zip ties.